The Natural Origins of Obligation

A friend of yours is moving to a fifth-floor walkup. You offer to help. When the day arrives – a beautiful Saturday – you realize that you’ve volunteered for a much bigger, sweatier, and uglier job than you anticipated. The urge to make some excuse and bail out is strong, but you push it back down. Leaving your friend in the lurch would make you feel terrible. So you sigh, pull on your oldest jeans and sneakers, and get ready for a long day. Where does this guilty sense that we ought to follow through on our obligations come from? The developmental psychologist Michael Tomasello, one of the leading figures in human evolutionary science, argues in a new paper that it hinges on our rare ability to think of ourselves as part of a collective “we.”

Obligation isn’t empathy (feeling or intuiting the feelings of others) or sympathy (the motivation to act on feelings of empathy to relieve someone else’s distress). Sympathy functions in a positive way to directly motivate social connection, incentivizing us to perform acts of succor and aid. When a naturally sympathetic person sees a friend crying, he doesn’t feel obligated to put an arm around his friend: he feels the urge to. Obligation, on the other hand, is, in Tomasello’s words, a “stick, not a carrot:”  it motivates us in a negative way to perform acts that, all things being equal, we might often prefer not to do, in order to fulfill promises or to avoid feelings of guilt.

This feeling of obligation is probably unique to humans. It makes possible the complex web of roles and relationships that define our social lives, enabling us to live up to expectations and fulfill responsibilities that, if left to our own devices, we’d just as soon avoid, but which others depend on. Fixing breakfast for one’s kids every morning, showing up to work on time, obeying traffic laws,* refusing bribes,† writing tedious case reports – these are all examples of things we do not out of sympathy, but out of obligation. 

Obligation and Agreement

Tomasello points out that, also unlike sympathy, obligation always implicates an agreement or promise — implied or explicit — between at least two people. You wouldn’t feel obligated to spend your Saturday lugging heavy objects around if you hadn’t agreed to help your friend in the first place. Sure, maybe a sudden rush of sympathy prompted you to blurt out your initial offer to help. But it was the agreement itself, not the sentiments underlying it, that created the obligation. And now that the obligation exists, gushy sentiments and feelings aren’t enough to hold up your end of the bargain. You actually have to follow through on your promise, regardless of how sweaty and sore you might be afterward — or you’ll risk a major breach in your friendship.

By contrast, we don’t need a preexisting relationship or agreement to feel sympathy for someone. In fact, we don’t even need to know who the other person is. Think of those mail solicitations from international charities that use closeup photographs of big-eyed, underprivileged children to tug at our heartstrings. Those charities depend on powerful gushes of natural sympathy stirring people to whip out their checkbooks and relieve total strangers’ suffering.

In the same vein, soldiers from opposing armies can even feel human sympathy for each other in spite of their nominal obligation to treat each other as enemies. In the movie Hacksaw Ridge, for example, the protagonist Desmond Doss, a medic and conscientious objector, covertly bandages a wounded Japanese soldier’s injuries on Okinawa Island – an event that probably had a basis in fact.

So sympathy gets a lot of good press. But obligation is much rarer, biologically speaking. Plenty of other animals can feel and act on sympathy, from dogs who comfort their owners to rats who show distress when they see other rats in trouble, yet obligation is limited to the world of human beings. This uniqueness makes obligation an enticing puzzle. 

A Shared “We”

Tomasello argues that obligation emerges fundamentally from the shared sense of an interdependent “we,” which entails a shared intentionality to collaborate together toward a common goal, often with distinct roles and responsibilities for different partners. For example, two people moving a sofa up a flight of stairs act as a “we” in this way. They’re adjusting their plans according to each others’ actions, envisioning and compensating for each others’ limited point of view (“Watch out – there’s another step behind you!”), making decisions that complement and are constrained by the other person. Moving the sofa is something they both intend to do, and is something they intend to do together. 

In this context of shared intentionality, it would be a major no-no to suddenly drop your end of the sofa and depart for lunch without warning just because you felt hungry, leaving your friend stranded on the staircase. Shared intentionality creates mutual obligations that are “sticky” – once you’ve entered into them, others expect you to persist in them until the collective “we” agrees that it’s time to stop. 

For example, maybe the sofa gets stuck on the last flight of stairs, and after struggling with it for a while, you both look at each other, realize you need a break, and climb back downstairs to get some Chinese food. In this scenario, the joint endeavor is over (or temporarily paused) by agreement, so it’s okay to walk away. It doesn’t have to be a verbal agreement – in studies of children’s sense of obligation, Tomasello and others have found that kids often ask “permission” to leave a game simply by looking anxiously at the grownups. The point is that both toddlers and adults understand that, once you’ve entered into a joint endeavor with others, it’s a breach of etiquette to simply exit the agreement at will. 

A key mechanism behind this “stickiness” is what Tomasello calls partner control. At some point during human evolution, cooperation became the all-important strategy. Food-sharing and collaborative hunting meant that if you didn’t have people to cooperate with you, you died. (This isn’t true of chimps and other great apes, who, while being highly social, are usually capable of foraging enough food for themselves.) The resultant need for reliable comrades put pressure on individuals to themselves become more reliable, and so boost the probability that promising coalitions would choose them to be members. In other words, partner choice – the ability for individuals to choose who they’d collaborate with – put pressure on everyone to be the best collaboration partner they could be.

Partner control is the next step up from partner choice. In partner control, people aren’t just picky about who they cooperate with – they also try to improve the partners they already have. The vast communicative flexibility of language makes it possible for human beings to protest, challenge, or complain about the particulars of someone else’s bad behavior, transforming the strategic cooperative relationship into one based on negotiation – if I complain that you didn’t help out during the hunt, you can apologize and promise to behave better in the future, and so repair the relationship. Apologizing serves to acknowledge the legitimacy of the obligation.

Guilt Vs. Obligation

Internal feelings, such as of guilt or obligation, then come to represent these social relationships and negotiations within the minds of each participant. We introject the perspective of the shared “we,” regulating our behavior in accordance with the roles we occupy and those shared goals that we recognize as legitimate. Self-regulation, in this view, is really group or social regulation, projected inward. If we don’t “partner control” ourselves – adjusting our own behavior to fit in line with how we know that desirable cooperation partners ought to behave – then we risk being cast out, isolated. 

But the key is that everyone is doing this simultaneously, each person trying to self-regulate in accordance with the expectations of the rest of the coalition’s members. And we’re not doing it dyadically. I don’t just adjust my behavior to fit your expectations or demands (or my internal guesses about what those are). Instead, we each adjust our behavior reciprocally to adapt to the shared expectations of the group.

This shift from a dyadic to a triadic relationship even affects which moral emotions we feel. We feel shame when our partners think ill of us directly, but breaches of obligation evoke guilt related to the shared “we.” This triadic structure makes guilt an inherently more complex emotion, reflecting a social whole that is greater than the sum of the parts.

Take the concept of property and ownership. Ownership – the exclusive right to possess and use an object – is ubiquitous across cultures, and imposes obligations and rights on people who recognize it. Yet, as Tomasello writes,

ownership is not just about how I relate to my iPhone, but rather about how you and I relate to one another with respect to my iPhone. …This triadic structure – you and I relating to one another about some external object or action – is the defining organization of social activities structured by shared intentionality.

Collective Intentionality

From shared intentionality, in turn, emerges collective intentionality, a kind of interdependent “we” writ large. For Tomasello, collective intentionality extends beyond small, interpersonal collaborations to allow all members of a broader society to recognize the same rules and laws – the sum of the general obligations that apply to everyone who is one of “us,” whether she’s a personal acquaintance or not. Collective intentionality is why most of us feel an obligation to stop at red lights even late at night, or why we expect firefighters to show up if we call in a blaze. Society itself is, from this perspective, a kind of huge joint endeavor: a massively scaled-up version of two people moving a couch. We self-regulate in order to meet the expectations of the collective “we.”

Of course, in real life we don’t always meet our obligations. People cheat on taxes because they want the extra money, or they come in late to work because they relished the extra sleep. But these failures are expected under Tomasello’s framework. If there were no temptation to skip out on our obligations, then we wouldn’t need all that internalized sense of social pressure and guilt to begin with. That is, the very existence of obligation implies, by definition, a conflict between social requirements and personal motives. This is what the sociologist of religion Émile Durkheim called homo duplex: the necessary, inescapable conflict between spontaneous, personal motives that meet immediate preferences or needs, and social obligations that make it possible for people to rely on each other over time.

Ultimately, according to Tomasello, the only obligations we actually strive to uphold are the ones that we consider legitimate. And legitimacy is fundamentally a social thing, since we have to agree – implicitly or not – on the mutual responsibilities that attend a relationship in order for that relationship to enjoy legitimacy. This is why most forms of slavery wouldn’t fit in this scheme of obligation – simply forcing someone to work for you doesn’t create legitimacy or a sense of mutual obligation, because it doesn’t entail a shared “we.” It’s also why people who reject the legitimacy of their wider culture often act in ways that seem antisocial or rebellious – to them, the collective “we” is illegitimate, and so any obligations to it are void. 

With moral psychology enjoying plenty of attention in our rollicking era of political weirdness and cascading social uncertainty, plenty of high-profile scholars have taken stabs at trying to explain morality in purely naturalistic terms. For Oliver Scott Curry, morality boils down to cooperation, full stop. For Jonathan Haidt, morality comes in different “flavors,” with conservatives preferring authority and loyalty to liberals’ fairness and harm prevention. Tomasello’s work on obligation points in a different direction, illuminating an oft-shadowed corner of the moral kitchen, but one where a lot of, maybe even most of, the most important cooking gets done.

* Except in Boston and Miami

† May not apply in Russia